Freescale Semiconductor, Inc. Order this document by AN827/D SEMICONDUCTOR TECHNICAL DATA ! # "!! Prepared by: PLL Applications The theory in this application note is still applicable, but some of the products referenced may be discontinued. Freescale Semiconductor, Inc... INTRODUCTION The MC12009, MC12011, or MC12013 can be used as part of a variable modulus (divisor) prescaling subsystem used in certain Digital Phase–Locked Loops (PLL). More often than not, the feedback loop of any PLL contains a counter–divider. Many methods are available for building a divider, but not all are simple, economical, or convenient in a particular application. The technique and system described here offer a new approach to the construction of a phase–locked loop divider. In addition to using either the MC12009, MC12011, or the MC12013 variable modulus prescaler, this system requires an MC12014 Counter Control Logic Function, together with suitable programmable counters (e.g., MC4016s or SN74LS716s). Data sheets for these additional devices should be consulted for their particular functional descriptions. quency) to give a new reference frequency: channel spacing/P. Another solution is found by considering the defining Equation 1 for fout of Figure 1. From the equation it may be seen that only every P channel can be programmed simply, because N is always an integer. To obtain intermediate channels, P must be multiplied by an integer plus a fraction. This fraction would be of the form: A/P. If N is defined to be an integer number, NP, plus a fraction, A/P, N may be expressed as: N = NP + A/P. Substituting this expression for N in Equation 1 gives: or fout = (NP + A/P) • P • fref fout = (NP P + A) • fref fout = N • P • fref (1) where P is fixed and N is variable. For a change of 1 in N, the output frequency changes by P • f ref. If f ref equals the desired channel spacing, then only every P channel may be programmed using this method. A problem remains: how to program intermediate channels. One solution to this problem is shown in Figure 2. A ÷ P is placed in series with the desired channel spacing (fre- (3) fout = NP • P • fref + A • fref. (4) Equation 4 shows that all channels can be obtained directly if N can take on fractional values. Since it is difficult to multiply by a fractional number, Equation 4 must be synthesized by some other means. fout = N • P • fref DESIGN CONSIDERATIONS The disadvantage of using a fixed modulus (÷ P) for frequency division in high–frequency phase–locked loops (PLL) is that it requires dividing the desired reference frequency by P also (desired reference frequency equals channel spacing). The MC12009/11/13 are especially designed for use with a technique called “variable modulus prescaling”. This technique allows a simple MECL two–modulus prescaler to be controlled by a relatively slow MTTL programmer counter. The use of this technique permits direct high–frequency prescaling without any sacrifice in resolution since it is no longer necessary to divide the reference frequency by the modulus of the high–frequency prescaler. The theory of “variable modulus prescaling” may be explained by considering the system shown in Figure 1. For the loop shown: (2) REFERENCE FREQUENCY (fref) PHASE DET. F (s) VCO PROGRAMMABLE DIVIDER ÷N ÷P Figure 1. Frequency Synthesis by Prescaling fout = N • fref REFERENCE FREQUENCY (fref) ÷P PHASE DET. F (s) PROGRAMMABLE DIVIDER ÷N VCO ÷P Figure 2. Frequency Synthesis by Prescaling ORIGINALLY PRINTED 5/81 REFORMATTED 10/95 Motorola, Inc. 1995 MOTOROLA For More Information On This Product, Go to: www.freescale.com AN827 1 Freescale Semiconductor, Inc. Taking Equation 3 and adding ± AP to the coefficient of the fref, the equation becomes: fout = (NP • P + A + A • P – A • P) fref. (5) Collecting terms and factoring gives: fout = [(NP – A) P + A (P + 1)] fref (6) From Equation 6 it becomes apparent that the fractional part of N can be synthesized by using a two–modulus counter (P and P + 1) and dividing by the upper modulus, A times, and the lower modulus (NP – A) times. Equation 6 suggests the circuit configuration in Figure 3. The A counter shown must be the type that counts from the programmed state (A) to the enable state, and remains in this state until divide by NP is completed in the programmable counter. Freescale Semiconductor, Inc... fout = Nfref REFERENCE FREQUENCY (fref) PHASE DET. F (s) VCO TWO–MODULUS PRESCALER P, P + 1 PROGRAMMABLE COUNTERS ÷ NP PROGRAMMABLE COUNTERS ÷A N = NP • P + A Figure 3. Frequency Synthesis by Two–Modulus Prescaling In operation, the prescaler divides by P + 1, A times. For every P + 1 pulse into the prescaler, both the A counter and NP counter are decremented by 1. The prescaler divides by P + 1 until the A counter reaches the zero state. At the end of (P + 1) • A pulses, the state of the NP counter equals (NP – A). The modulus of the prescaler then changes to P. The variable modulus counter divides by P until the remaining count, (NP – A) in the NP counter, is decremented to zero. Finally, when this is completed, the A and NP counters are reset and the cycle repeats. To further understand this prescaling technique, consider the case with P = 10. Equation 6 becomes: fout = (A + 10 NP) • fref If NP consists of 2 decades of counters then: (7) NP = 10 NP1 + NP0 (NP1 is the most significant digit), and Equation 7 becomes: fout = (100 NP1 + 10 NP0 + A) fref. To do variable modulus prescaling using the variable modulus prescalers (MC12009/11/13) and programmable divide by N counters (MC4016, MC4018) one additional part is required: the MC12014 (Counter Control Logic). In variable modulus prescaling the MC12014 serves a dual purpose: it detects the terminal (zero) count of the A counter, AN827 2 to switch the modulus of the MC12013; and it extends the maximum operating frequency of the programmable counters to above 25 MHz. (See the MC12014 data sheet for a detailed description of the Counter Control Logic). Figure 4 shows the method of interconnecting the MC12013, MC12014, and MC4016 (or MC4018) for variable modulus prescaling. To understand the operation of the circuit shown in Figure 4, consider division by 43. Division by 43 is done by programming NP1 = 0, NP0 = 4, and A = 3. Waveforms for various points in the circuit are shown in Figure 5 for this division. From the waveforms it may be seen that the two–modulus prescaler starts in the divide by 11 mode, and the first input pulse causes point A to go high. This positive transition decrements the NP counter to 3, and counter A to 2. After 11 pulses, point A again goes high; the NP counter decrements to 2 and the A counter to 1. The “2” contained in the NP counter enables the inputs to the frequency extender portion of the MC12014. After 11 more pulses point A goes high again. With this position transition at A , the output (fout) of the MC12014 goes low, the NP counter goes to 1, and the A counter goes to 0. The zero state of the A counter is detected by the MC12014, causing point B to go to 1 and changing the modulus of the MC12013 to 10 at the start of the cycle. When fout goes low, the programmable counters are reset to the programmed number. After 11 pulses (the enable went high after the start of the cycle and therefore doesn’t change the modulus until the next cycle), point A makes another positive transition. This positive transition causes f out to return high, release the preset on the counter, and generates a pulse to clear the latch (return point B to 0). After 10 pulses the cycle begins again (point B was high prior to point A going high). The number of input pulses that have occurred during this entire operation is: 11 + 11 + 11 + 10 = 43. Figures 6 and 7 show the waveforms for divide by 42 and divide by 44 respectively. The variable modulus prescaling technique may be used in any application as long as the number in the NP counter is greater than or equal to the number in the A counter. Failure to observe this rule will result in erroneous results. (For example, for the system shown in Figure 4, if the number 45 is programmed, the circuit actually will divide by 44. This is not a serious restriction since NP is greater than A in most applications.) It is important to note that the A counter has been composed of only one counter for discussion only; where required, the A counter may be made as large as needed by cascading several programmable counters. Figure 8 shows the method of interconnecting counters. Operation is previously described. The number of stages in the A Counter should not exceed the number of stages for the NP counters. As many counters as desired may be cascaded, as long as fan–in and fan–out rules for each part are observed. The theory of “variable modulus prescaling” developed above, examined a case in which the upper modulus of the two–modulus prescaler was 1 greater than the lower modulus. However, the technique described is by no means limited to this one special case. There are applications in which it is desirable to use moduli other than P/(P + 1). For More Information On This Product, Go to: www.freescale.com MOTOROLA Freescale Semiconductor, Inc. It can be shown that for a general case in which the moduli of the two–modulus prescaler are P and P + M, Equation 6 becomes: From Equation 8 it may be seen that the upper modulus of the two–modulus prescaler has no effect on the NP counter, and that the number programmed in the A counter is simply multiplied by M. There is no one procedure which will always yield the best counter configuration for all possible applications. Each designer will develop his own special design for the counter portion of his PLL system. fout = [(NP – A) P + A (P + M)] • fref or fout = [NP • P + M • A] • fref. (8) Z0 Z1 Freescale Semiconductor, Inc... Z2 A E fin C1 Eo Z3 MTTLout MC12013 fin P0 B MC12014 P1 C P2 fout fout P3 B1 Q0 Q1 Q2 Q3 Q0 Q1 Q2 Q3 C MC4018 C MC4016 PE A B C MC4018 PE NP0 R PE NP1 Figure 4. Direct Programming Utilizing Two–Modulus Prescaler MOTOROLA For More Information On This Product, Go to: www.freescale.com AN827 3 Freescale Semiconductor, Inc. 11 PULSES 11 PULSES 11 PULSES 10 PULSES 11 PULSES 11 PULSES 11 PULSES fin A B 1 CYCLE = 43 PULSES C fout Freescale Semiconductor, Inc... Figure 5. Waveforms for Divide by 43 11 PULSES 11 PULSES 10 PULSES 10 PULSES 11 PULSES 11 PULSES 10 PULSES fin A B 1 CYCLE = 42 PULSES C fout Figure 6. Waveforms for Divide by 42 11 PULSES 11 PULSES 11 PULSES 11 PULSES 11 PULSES 11 PULSES 11 PULSES fin A B 1 CYCLE = 44 PULSES C fout Figure 7. Waveforms for Divide by 44 AN827 4 For More Information On This Product, Go to: www.freescale.com MOTOROLA Freescale Semiconductor, Inc. Z0 Z1 Z2 Eo Z3 B2 E fin C1 MTTLout MC12013 fin P0 MC12014 P1 P2 fout fout P3 Freescale Semiconductor, Inc... B1 Q0 Q1 Q2 Q3 MC4016 C PE A0 Q3 B MC4016 R PE A1 A COUNTER A = 100 A2 + 10 A1 + A0 C Q3 MC4016 B PE A2 C MC4016 B PE Q0 Q1 Q2 Q3 MC4016 C PE Q3 B MC4016 R C PE NP0 NP1 C Q3 MC4016 B PE C Q3 MC4016 B PE NP2 NP COUNTER NP = 100 NP2 + 10 NP1 + NP0 Figure 8. Method of Interconnecting Counters MOTOROLA For More Information On This Product, Go to: www.freescale.com AN827 5 Freescale Semiconductor, Inc. 50 kHz REFERENCE FREQUENCY PHASE DET. MC4044 F (s) Eout MC12014 fout B1 P3 P2 P1 P0 B2 Z3 fout = 100 TO 200 MHz IN 50 kHz STEPS VCO MC1648 Z2 1/2 MC3060 C Q Z1 Z0 Q ÷ 10/÷ 11 C D MTTLout MC12013 Freescale Semiconductor, Inc... ÷ 20/÷ 21 B R MC4016 Q3 C B Q3 Q2 Q1 MC4016 C MC4016 B Q0 C PE D3 D2 D1 D0 PE D3 D2 D1 D0 PE D3 D2 D1 D0 100 MHz 10 MHz 1 MHz Q0 R MC4016 C PE D3 D2 D1 D0 MC4016* C PE D3 D2 D1 D0 100 kHz 50 kHz * Used as Two–Modulus Programmable Counter. Also could use MC4017. Figure 9. Direct Programming 100 – 200 MHz Synthesizer in 50 kHz Steps Motorola reserves the right to make changes without further notice to any products herein. Motorola makes no warranty, representation or guarantee regarding the suitability of its products for any particular purpose, nor does Motorola assume any liability arising out of the application or use of any product or circuit, and specifically disclaims any and all liability, including without limitation consequential or incidental damages. “Typical” parameters can and do vary in different applications. All operating parameters, including “Typicals” must be validated for each customer application by customer’s technical experts. Motorola does not convey any license under its patent rights nor the rights of others. Motorola products are not designed, intended, or authorized for use as components in systems intended for surgical implant into the body, or other applications intended to support or sustain life, or for any other application in which the failure of the Motorola product could create a situation where personal injury or death may occur. Should Buyer purchase or use Motorola products for any such unintended or unauthorized application, Buyer shall indemnify and hold Motorola and its officers, employees, subsidiaries, affiliates, and distributors harmless against all claims, costs, damages, and expenses, and reasonable attorney fees arising out of, directly or indirectly, any claim of personal injury or death associated with such unintended or unauthorized use, even if such claim alleges that Motorola was negligent regarding the design or manufacture of the part. Motorola and are registered trademarks of Motorola, Inc. Motorola, Inc. is an Equal Opportunity/Affirmative Action Employer. How to reach us: USA/EUROPE: Motorola Literature Distribution; P.O. Box 20912; Phoenix, Arizona 85036. 1–800–441–2447 JAPAN: Nippon Motorola Ltd.; Tatsumi–SPD–JLDC, Toshikatsu Otsuki, 6F Seibu–Butsuryu–Center, 3–14–2 Tatsumi Koto–Ku, Tokyo 135, Japan. 03–3521–8315 HONG KONG: Motorola Semiconductors H.K. Ltd.; 8B Tai Ping Industrial Park, 51 Ting Kok Road, Tai Po, N.T., Hong Kong. 852–26629298 AN827 6 ◊For More Information On This Product, Go to: www.freescale.com *AN827/D* AN827/D MOTOROLA